A genome is the total DNA contained in one cell, containing the entirety of the genes and the intergenic regions of a living organism, and organisms known to us are defined by the genome. Currently, reading of the human nucleotide sequences has already been completed, and the number of genes is being predicted. However, nucleotide sequence information is not sufficient to elucidate biological phenomena, and although the total genomes of Escherichia coli, yeast, Caenorhabditis elegans, drosophila and the like have actually been elucidated, it does not mean that the entirety of the gene function of these model living organisms has been elucidated.
Proteomics is an important item in the post-genomic era, and this technology is gathering attention from the perspective of carrying out functional analysis of genes. Owing to the current proteome technology, that is to say, a mass spectroscopic device, a database, and a search engine linking these, the ease with which a protein is identified has leaped forward, allowing a specific protein mixture (for instance, a protein complex) to be identified comprehensively (for example, see Non-patent References 1 to 4). However, merely identifying a protein, or merely elucidating a component, is insufficient as functional analysis. The reason is, a proteome being dynamic, the expression state thereof varies depending on the timing or the location. Therefore, carrying out an analysis of the variation of an individual protein should allow the advantages of proteomics to be exploited. Thus, similarly to Differential Display in a DNA chip, a Differential Display at the protein level is created also in proteome, and a variety of attempts to capture the variation of individual protein level are carried out.
Two-dimensional electrophoresis is routinely used as means for proteome analysis; however, with such a method that compares spots of protein stained on a gel, comparing accurately the quantity of individual protein present in two samples to be compared is extremely difficult. The reasons being, issues exist in:    1. reproducibility of extraction from cell and reproducibility due to pretreatment manipulation such as cell fractionation/crude purification    2. reproducibility of protein separation in two-dimensional electrophoresis and the like    3. reproducibility of gel staining and the like; furthermore,    4. when a plurality of proteins are mixed in one spot/band, the quantitative aspect with respect to an individual protein is lost.Thus, currently, capturing a small variation is difficult with a pattern comparison by gel staining, except when the expression amount varies dramatically, i.e., 10 to 100 times or more.
In addition, even if a dramatic variation is detected in this way, by merely comparing a spot on a gel, the protein corresponding to this spot remains unidentified, and the target spot must be cut out from the gel and protein identification work be carried out with a mass spectroscopic device.
Obtaining quantitative data in an analysis by the mass spectroscopic device necessarily requires correction with an internal standard substance, as reproducibility of ionization rate is poor. In addition, the quantitative value varies exceedingly for each experiment due to the pretreatment procedure (extraction, derivatization, separation and the like) up to the analysis with the mass spectroscopic device. Thus, by pre-adding to a sample an internal standard substance obtained by labeling with a stable isotope a target low molecular weight substance, protein or the like, all these corrections has become possible, and improvement of quantitation accuracy has kept forward. Quantitative analysis is carried out in a drug, a lipid, a protein or the like, using an internal standard substance labeled with a stable isotope element (for example, see Non-patent References 5 to 10).
The aforementioned quantitation of a low molecular substance such as a drug or a lipid, and protein by the mass spectroscopic device is applicable only when the target substance is known. Meanwhile, in proteomics, unidentified and numerous target substances must be measured. Therefore, in recent years, quantitative proteome using the mass spectroscopic device has become possible by labeling the entirety of the proteins contained in the samples to be compared. Currently reported methods (for example, see Non-patent Reference 11) can be divided roughly into the following three classes (see FIG. 1):
In Vivo Labeling Method
As a quantitative proteome that combines a stable isotope and a mass spectroscopic device, a method by Oda et al. exists, in which yeast proteins are labeled with 15N (for example, see Patent Reference 1 and Non-patent References 12 to 13). Two types of cells to be compared are cultured respectively in a culture medium containing an isotope labeled amino acid and a culture medium not containing an isotope labeled amino acid, the cultured cells are mixed, then proteins are extracted, fractionated, and analyzed with the mass spectroscopic device, and the peak intensities of isotope labeled proteins and isotopically unlabeled proteins are compared. With this method, even if there is an error in the recovery rate during the analysis, the intercellular difference can be obtained quantitatively, and in addition, identification of the proteins can be carried out simultaneously with the spectrum data.
However, while this in vivo labeling method is in principle applicable to every cell as long as they are cultured cells, in reality, such as large change in cell growth rate due to the change in culture conditions, culture in a culture medium containing a stable isotope is often difficult. In other words, it can be performed only with limited cells, which are not influenced by the culture condition. In addition, tissue from animal and human or the like, cannot be cultured easily, and cannot be measured with this method.
Method for In Vitro Labeling Before Digestion
Thus, the following method has been found, as a method for comprehensively quantitating proteins in a tissue. That is to say, protein quantitation became possible by adding an isotope labeled reagent reactive to cysteine residues and an non-isotope labeled reagent reactive to cysteine residues respectively to two types of samples to be compared and labeling cysteine residues on proteins contained in the samples, purifying the proteins with a column, analyzing with a mass spectroscopic device, and comparing a pair of peak intensities (for example, see Non-patent References 14 to 19). However, proteins that do not contain a cysteine residue cannot be quantified with this method, in addition, the variation of quantitative value at each experiment in the pretreatment stage such as purification cannot be corrected.
Method for In Vitro Labeling after Digestion and During Digestion
Some methods for proteins not containing a cysteine residue and some methods with improved purification operability have been reported. Protein quantitation became possible by digesting via an enzymatic reaction a protein contained in two types of samples to be compared, labeling the obtained C-terminus and N-terminus of a peptide fragment, glutamic acid residue, aspartic acid residue and the like respectively with an isotope labeled molecule and a non-isotope labeled molecule, analyzing with a mass spectroscopic device, and comparing a pair of peak intensities (for example, see Non-patent Reference 20 to 23).
However, with these methods, the variation of quantitative value at each experiment in the pretreatment stage, such as digestion, fractionation and purification of the protein contained in the sample, cannot be corrected.
As described above, current quantitative proteome analysis method can be divided roughly into in vivo labeling method and in vitro labeling method. In the former, labeling is simple, and highly reproducible data can be expected; however, when the sample is a tissue, a biological fluid, a cell organ, a protein complex, a cell incapable of growing in a culture medium containing a stable isotope, or the like, it cannot be applied to proteins that do not incorporate stable isotope contained therein. The latter can be applied to every sample; however, the variation of quantitative value at each experiment in the pretreatment stage until labeling, such as sample disruption, extraction, digestion, fractionation and purification, cannot be corrected. In addition, the labeling reaction per se is often not simple such that differences in the experience of the experimenters show.
Quantitative proteome analysis methods thus far has been directly comparing an internal standard substance and a test sample, such that sample preparation needed to be carried out for each measurement. Further, carrying out comparison of three or more samples was not simple, which also decreased accuracy. Furthermore, when the measurement subject changed the internal standard substance to be labeled also changed, each time requiring the labeling conditions to be examined; in addition, comparison between experiments was not possible.
Moreover, quantitative proteome analysis method thus far has been directly comparing an internal standard substance and a test sample, and when the measurement subject changed, the internal standard substance to be labeled also needed to be changed. On the other hand, quantitating the internal standard substance biological molecule was not simple; therefore, absolute quantitation was very difficult.
Among the quantitative proteome analysis methods reported thus far, the AQUA method exists as an absolute quantitation method (for example, see Patent Reference 2). This is a method in which a stable isotope is incorporated into a portion of a synthetic peptide, and determines the ratio with respect to the target protein. However, with this method, an isotope labeled synthetic peptide must always be prepared for each experiment, quantification must also be carried out at each experiment, such that comprehensive quantitation is difficult. In addition, it is influenced by the recovery rate in the purification process of the objective protein, and also influenced by the efficiency of in-gel digestion generally told to be 30-50%, such that accurate quantitation is difficult.
On the other hand, as we stepped into the post genomic era, it has become more and more important to separate, identify and further quantitate proteins, which are important biological molecules in a living organism. In particular, for research and development of techniques for diagnosis and treatment of diseases, the function of multiple proteins must be elucidated.
Conventionally, two-dimensional electrophoresis has been used for the proteome analysis of proteins expressed by a cell. Here, the term “proteome analysis” means analysis for elucidating the relationship between genetic information and various proteins interacting intricately inside the cell (for example, see Non-patent Reference 24). In other words, proteome analysis designates a method for comprehensively analyzing all the proteins constituting a cell.
In the above two-dimensional electrophoresis, the expressed proteins are developed on a gel, and the spot corresponding to the subject protein is cut out to identify the type thereof comprehensively. Therefore, two-dimensional electrophoresis is a qualitative analysis means that is useful in proteome analysis.
However, it has been pointed out that two-dimensional electrophoresis was not suitable for a quantitative analysis of protein owing to the quantity of protein developed being very small and the recovery rate at analytical time being prone to errors.
Meanwhile, mass spectroscopy is used as another important protein analysis technique. The present method is a method for analyzing an accurate mass of a protein or a peptide using a mass spectroscopic device. This mass spectroscopic device generally comprises a device for ionizing proteins and peptides, and a mass separation unit for separation according to their masses, and is constituted by a mass spectrometer for analyzing the mass. Then, the ease with which a protein is identified today can be said to have kept forward with a mass spectrometer, a protein database, and a search system linking these. Therefore, a specific protein group (for example, a group of proteins that form a given complex) can be comprehensively identified.
However, merely identifying a protein, or elucidating components by mass spectroscopy is insufficient as functional analysis of protein. The reason being, in a functional analysis by proteome, comparison with a pathology, and analysis of the variation of an individual protein due to the influenced of a stimulation by a drug or a ligand, knock out or overexpression of a target gene, and the like, that is to say, quantitative data allowing the amount of protein expressed to be compared, become necessary.
Based on such a background, quantitative analyses using a mass spectroscopic device exploiting isotope labeling have been attempted (for example, see Non-patent Reference 25).    [Non-patent Reference 1] “Nature Genetics” Volume 20, September 1998, p. 46-50    [Non-patent Reference 2] “Journal of Cell Biology” Volume 141, No. 4, 18 May 1998, p. 967-977    [Non-patent Reference 3] “Current Opinion in Cell Biology” Volume 15, No. 2, April 2003, p. 199-205    [Non-patent Reference 4] “Current Opinion in Chemical Biology” Volume 7, No. 1, February 2003, p. 21-27    [Non-patent Reference 5] “Surgery Today” Volume 23, No. 3, 1993, p. 228-233    [Non-patent Reference 6] “Rapid Communications in Mass Spectrometry” Volume 8, No. 5, 1994, p. 377-380    [Non-patent Reference 7] “Journal of Chromatography A” Volume 694, No. 1, May 1995, p. 209-218    [Non-patent Reference 8] “Analytical Biochemistry” Volume 231, No. 1, October 1995, p. 141-150    [Non-patent Reference 9] “Journal of Chromatography B” Volume 729, Nos. 1-2, June 1999, p. 147-155    [Non-patent Reference 10] “Proceedings of the National Academy of Sciences of the United States of America” Volume 100, No. 12, June 2003, p. 6940-6945    [Non-patent Reference 11] “Current Opinion in Chemical Biology” Volume 7, No. 1, February 2003, p. 70-77    [Non-patent Reference 12] “Proceedings of the National Academy of Sciences of the United States of America” Volume 96, No. 12, June 1999, p. 6591-6596    [Non-patent Reference 13] “Methods” Volume 29, No. 2, February 2003, p. 124-130    [Non-patent Reference 14] “Nature Biotechnology” Volume 17, No. 10, October 1999, p. 994-999    [Non-patent Reference 15] “Analytical Chemistry” Volume 70, No. 24, December 1998, p. 5150-5158    [Non-patent Reference 16] “Rapid Communications in Mass Spectrometry” Volume 16, No. 15, 2002, p. 1416-1424    [Non-patent Reference 17] “Journal of American Society of Mass Spectrometry” Volume 12, No. 12, December 2001, p. 1238-1246    [Non-patent Reference 18] “Nature Biotechnology” Volume 19, No. 10, October 2001, p. 946-951    [Non-patent Reference 19] “Molecular and Cellular Proteomics” Volume 1, No. 1, January 2002, p. 19-29    [Non-patent Reference 20] “Analytical Chemistry” Volume 73, No. 13, July 2001, p. 2836-2842    [Non-patent Reference 21] “Rapid Communications in Mass Spectrometry” Volume 15, No. 14, 2001, p. 1214-1221    [Non-patent Reference 22] “Nature Biotechnology” Volume 20, No. 2, February 2002, p. 163-170    [Non-patent Reference 23] “Nature Biotechnology” Volume 20, No. 5, May 2002, p. 512-515    [Non-patent Reference 24] Karn, P. Science 270, pp. 369-370, 1995    [Non-patent Reference 25] “Proteome Analysis Method”, Yodosha, printed 10 Jul. 2000, pp. 111-122    [Patent Reference 1] U.S. Pat. No. 6,391,649    [Patent Reference 2] WO2003 016861